Customer Behavior Prediction for E-Commerce Sites Using Machine Learning Techniques

Abstract

Every business organization depends on the intelligent decision analytical system to analyze consumer behavior. This analysis and prediction can impact mainly the demand-driven supply chain management of the organization. Data analysts use different techniques like data mining and machine learning (ML) approaches to discover the hidden patterns in consumer behavior and thereby predict sales. As traditional methods of analysis cannot withstand the velocity of data collected by existing e-commerce sites, the newest approach using data mining and machine learning paved the best way for extracting hidden layers. This chapter is based on the analysis of customer behavior which is very useful to forecast next basket predictions, willingness to buy a product and personalized list suggestions which all aim at the satisfaction of the customer and making them stay loyal to a particular e-shopping site. This chapter reviews the methodologies used in earlier works on consumer behavior forecasting using techniques such as random forest, support vector machine, logistic regression, decision trees and neural networks. The result compares the performance of the ML algorithms and selects the logistic regression as the best modeling algorithm.